STATUS OF THE SP.-E. Bernaudin*, R

P. BoslanG. Olry,

Y. Gómez Martínez, LP

Abstract SPIRAL2 is a radioactive beams facility, c

superconducting linac driver, delivering deutenergy up to 40 MeV (up to 5 mA) and hean energy up to 14.5 MeV/u (up to superconducting linac is composed of twquarter wave resonators: type A (optimized 1 per cryomodule) and B (beta=0.12, 2 per The accelerator is scheduled to be commimid-2011 onwards. The project is therefore phase. This paper summarizes the latest restatus of the superconducting linac. All 16 tyhave been tested. Cryomodules from bothpresently being assembled in series. Instacryomodules in the new building in GANIL August 2011.

INTRODUCTION The GANIL's SPIRAL2 Project [1] aims

high intensities of rare isotope beams by adoproduction method for each respective radioa

The unstable beams will be produced b“Isotope Separation On-Line”, method via aby direct irradiation of fissile material. Onreferee reports of international experts andthe positive evaluations by IN2P3/CNRS anGANIL, and the support of the regioNormandie, the French Minister of Reseadecision on the construction of SPIRAL2 in

The driver will accelerate protons (0.133 MeV), deuterons (0.15 to 5 mA – 40 Meions (up to 1 mA, Q/A=1/3 – 14.5 MeV/u – 8.5 MeV/A). It consists of high perforsources, a RFQ, and the superconducting (Sdriver is also asked to provide all the e2 MeV/u to the maximum designed value.

The SC linac is composed of cryomoduleby CEA Saclay, and cryomodules B develOrsay. Both types of cavities are equipped wpower coupler specified for a maximum powCW (in travelling wave), developed in a thiLPSC Grenoble [2].

General development programs are quitboth cryomodules: a first qualification crybeen tested before the series. These cryomodules will be used in the machcomponents of the series (cavities and cryofabricated in industry. Cavities chemical trerinsing in clean room, assembly, and RF

PIRAL 2 SUPERCONDUCTING LINR. Ferdinand, SPIRAL2/GANIL Caen, France nd, CEA Saclay Irfu/SACM, France

IPN, CNRS/IN2P3, Orsay, France PSC, UJF, CNRS/IN2P3, INPG, Grenoble, Fran

composed of a terons with an avy ions with 1 mA). The

wo families of for beta=0.07, cryomodule).

issioned from in production

esults and the ype B cavities h families are allation of the

shall begin in

s at delivering opting the best active beam. by the ISOL, a converter, or n the basis of d committees, nd DSM/CEA, on of Basse-arch took the May 2005. 5 to 5 mA –

eV) and heavy to Q/A=1/6 –

rmance ECR SC) linac. The energies from

s A developed loped by IPN with the same wer of 40 kW ird laboratory,

te similar for yomodule has

qualification hine. All the omodules) are atments, HPR

F tests of the

cavities in vertical cryostat and RF cryomodules are performed in the resp

CRYOMODULEThe qualification cryomodule ha

between December 2008 and April after full disassembly, the qualificatiobeen tuned, HP rinsed, and reascryomodule. Tests of this module h2010 and are still ongoing.

In the meantime, the first two serieAS-3) have been received, one from e

Latest evelopments Early tests of the AZ-1 cavity sho

(10 times lower than expected) [3]. ThQ0 was caused by the dual purpose selinking the cavity to its bottom capdemonstrated by replacing these seal(see Fig. 1) during vertical cryostat these copper seals are replaced by Hel

Figure 1: Vertical tests results of the thTwo bottom curves (yellow and pinkcopper, dual purpose seal. All other cavities with an indium seal. The blucavity with an indium seal and copper

Tests in the qualification cryomoduvery difficult to maintain the niobium10K, mainly due to the thermal losscoupler. Therefore niobium caps havhigh RRR copper caps. Compuadditional RF losses would be 1 W inof RRR=70 copper. The AZ-1 cavity

NAC

nce

power tests of the pective labs.

E A as first been tested

2009 [3]. Recently, on cavity (AZ-1) has ssembled inside its have begun in April

s cavities (AZ-2 and each manufacturer.

wed a very poor Q0 he degradation of the eal (RF and vacuum) p. This failure was ls with indium ones tests. From now on licoflex© seals.

hree A-type cavities.

k) are cavities with a curves are the three

ue dot curve show a r bottom cap.

ule proved that it was m bottom cap below ses coming from the ve been replaced by utations show that n a bottom cap made has been tested with

D

MOPD025 Proceedings of IPAC’10, Kyoto, Japan

732

04 Hadron Accelerators

A08 Linear Accelerators

such a copper cap in vertical cryostat; results proved satisfactory (see Fig. 1).

Fine tuning of the cavities is performed chemically. Removing thickness preferentially at the top of the cavity (magnetic torus) lowers the frequency, whereas removing niobium preferentially at the bottom of the cavity (around the beam tubes) raises the frequency.

A new magnetic shield has been developed and manufactured. It is made of 1-mm thick Mumetall© plates assembled on the inside wall of the vacuum chamber. Its efficiency has been measured and is better than 50 around the cavity.

Ongoing Activities The qualification cryomodule is presently being tested

in Saclay. Cavity inside this cryomodule is the AZ-1 one, with a copper bottom cap fitted with a Helicoflex© seal. Preliminary results showed that alignment of the cavity inside the cryostat is unsatisfactory and shall be enhanced. We still have the possible adjustment of moving the whole cryomodule on axis at the price of a reduction of the apparent beam aperture. The cryogenic system shall also be slightly modified to limit apparent cryogenic losses and to stabilize the helium bath level; the real cryomodule consumption has still to be measured.

Coupler conditioning was performed easily both at 300 K and at 4 K. Conditioning procedure includes both pulsed and cw modes; pulsed modes are operated at 50Hz. Usual multipacting barriers have been observed at low power level (around 20 and 150 W of peak input power).

Figure 2: Series cavities just before tuning operation in the manufacturer premises.

A prototype LLRF system foreseen for the accelerator has been successfully tested. Despite high helium consumption (and with helium bath pressure fluctuations of ±2 mbar), the LLRF and tuning systems managed to maintain the phase shift within ±0.1° with respect to the RF pilot (specifications: ±0.5°), and the field amplitude in the cavity within ±0.12% (specifications: 1%).

The tuning system performed satisfactorily, until it blocked after 12,000 motor turns. The harmonic drive gear box, operating under vacuum and at 4 K, and the CuBe screw are under suspicion.

Further tests are being performed and full analysis of the results still has to be made.Series cavities fabrication is ongoing smoothly and according to schedule (see Fig. 2). All cavities are presently in the tuning stage. Tuning is performed by adjusting the cavity height: last two welds are the ones linking the stem to the top torus and the top torus to the cavity body respectively (as seen on Fig. 2). By cutting the stem and the cavity body to the proper length, it is possible to adjust the resonance frequency of the cavity with a sensitivity of 83 kHz/mm. This system has been qualified by both manufacturers on AZ-2 and AS-3 cavities. Manufacturers are required to deliver all cavities with a frequency precision of ±25 kHz.

Foreseen ctivities and chedule All elements of the first series cryomodule have been

delivered in Saclay. This cryomodule shall be assembled during the summer 2010 with the AZ-2 cavity. This cavity has already been tuned but shall be finally qualified in vertical cryostat before being put in the cryomodule. Tests of this cryomodule are scheduled for September 2010.

All series cavities shall be delivered in Saclay between June and September 2010. Upon reception they will be tuned chemically and individually tested in vertical cryostat for final acceptance. It is scheduled that all cavities will be qualified before the end of 2010. It will be the first final cryomodule.

In parallel, remaining cryomodules components, presently under fabrication, will be delivered before the end of the year in two or three batches. Assembly of the 10 remaining cryomodules shall begin before the end of 2010.

Final delivery of the last tested cryomodule to GANIL is foreseen for the first quarter of 2011.

CRYOMODULE B Cavities ests

The production of the 16 beta 0.12 cavities was achieved in November 2009 (see Fig. 3). All cavities have already been tested once in vertical cryostat at 4.5 K (see Fig. 4). There all satisfy the SPIRAL2 linac requirements (less than 10 W of dissipated power at 6.5 MV/m).

Figure 3: Family picture of some B cavities.

A S

T

Proceedings of IPAC’10, Kyoto, Japan MOPD025

04 Hadron Accelerators

A08 Linear Accelerators 733

Figure 4: Vertical tests results for all 16 typ

All cavities have been BCP etched. In omore homogenous etching rate, the chemicadone in two phases (2x2h, the cavity is tdown after the first phase) and the cavitdown” by circulating water within its hduring the process. We ended up with a +0.14 kHz/min, with a very good r(±0.02 kHz/min) for more than 15 chemical

The dissipated power at 6.5 MV/m has bea factor of two (mean value of 4 W instthanks to the baking of each cavity at 110°Cbaking is done in clean room after HPRdirectly blown in the stem and goes out bCF16 flanges of the helium vessel (see Figensures baking of the bottom end of the canot in contact with the helium vessel. Twrapped in a foil blanket during the process.

Figure 5: Sketch of the hot air flux for cav

Figure 6: MB14 cavity equipped with magnetic shield.

pe B cavities.

order to get a al treatment is turned upside ty is “cooled helium vessel sensitivity of

reproducibility processes.

een divided by tead of 8 W),

C for 48 h. The R. Hot air is by one of the g. 5). A heater avity, which is The cavity is .

vity baking.

its dual-layer

MagnThe two first magnetic shields hav

April 2010. The magnetic shield islayers of 1-mm thick Cryoperm© acooling circuit. The goal is to cool-d20 K) the magnetic shield before the transition, to ensure that the shieoptimal at the time of the transition. successfully tested with the MB14cryostat (see Fig. 6). The temperaturaround 30 K before the cavity transitio

Cryomodules Fabrication In addition to the prototype cryom

installed in the linac tunnel thanperformances [4], the productioncryomodules B components is going 2010. Three cryomodules have alreadIPN Orsay; the last one will be deliver

The assembly of the first cryomodcryomodule should be tested in Sepcryomodule will be individually testeevery 2 months, before shipment to G

RF POWER COUP21 of the 30 power couplers have b

LPSC laboratory in Grenoble. The ceramic window of two first c

during conditioning, seemingly becautoo thick. A thinner on (1nm) was tesgave higher multipacting. Thereforeabandoned for the remaining couplers

Of the remaining 19 delivered coufully tested and conditioned. No inccouplers are being delivered to the cryteams in Saclay and Orsay according t

Following the cryomodule A and B2010, the method to keep the temperwindow above 14°C (to avoid freeziTwo options are still being considregulated heater will be placed on theclose as possible to the ceramic, orflown in a channel around the ceramic

REFERENCES[1] T. Junquera, P. Bertrand, R

Jacquemet, “The high intensiLinac for the SPIRAL 2 projeLINAC 2006, Knoxville, USA, p

[2] Y. Gómez Martínez et al., “Texperimental result of the RF coSpiral2”, EPAC 06, Edinburgh, Ju

[3] P.-E. Bernaudin et al., “Tests statlow beta cryomodules”, PAC’02009, TU5PFP041.

[4] G. Olry et al., “SPIRAL2 cryomfirst results”, SRF09, Berlin, Sept

ve been delivered in s posed of two and a stainless steel down (ideally below normal to SC cavity

elding efficiency is One shield has been

4 cavity in vertical re of the shield was on.

module which will be nks to very good

n of the 6 other to be ended in June

dy been delivered to red in October 2010.

dule has begun. This ptember 2010. Each ed at IPN Orsay, one

GANIL.

PLERS been received by the

couplers broke down use TiN coating was sted with success by e TiN coating was

s. uplers, 6 have been cident occurred and yomodules assembly to schedule. B tests next autumn rature of the ceramic ing) will be chosen. dered: either a self e outer conductor as

r filtered air will be c.

S R. Ferdinand, M. ity superconducting ct at GANIL”, Proc.

142-144. heorical study and oupler prototypes of une 2006. tus of the SPIRAL 2 9, Vancouver, May

modules: status and tember 2009.

etic Shield

com

MOPD025 Proceedings of IPAC’10, Kyoto, Japan

734

04 Hadron Accelerators

A08 Linear Accelerators